devtools::install_github('charlie86/spotifyr')
## Skipping install of 'spotifyr' from a github remote, the SHA1 (ea9d985d) has not changed since last install.
## Use `force = TRUE` to force installation
library(spotifyr)
library(tidyverse)
## ── Attaching packages ────────────────────────────────── tidyverse 1.2.1 ──
## ✔ ggplot2 3.0.0 ✔ readr 1.1.1
## ✔ tibble 1.4.2 ✔ purrr 0.2.5
## ✔ tidyr 0.8.1 ✔ stringr 1.3.1
## ✔ ggplot2 3.0.0 ✔ forcats 0.3.0
## ── Conflicts ───────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
library(knitr)
library(beeswarm)
library(reshape2)
##
## Attaching package: 'reshape2'
## The following object is masked from 'package:tidyr':
##
## smiths
library(plotly)
##
## Attaching package: 'plotly'
## The following object is masked from 'package:ggplot2':
##
## last_plot
## The following object is masked from 'package:stats':
##
## filter
## The following object is masked from 'package:graphics':
##
## layout
devtools::install_github("josiahparry/geniusR")
## Skipping install of 'geniusR' from a github remote, the SHA1 (11fc6ec7) has not changed since last install.
## Use `force = TRUE` to force installation
library(geniusR)
##
## Attaching package: 'geniusR'
## The following objects are masked from 'package:spotifyr':
##
## add_genius, gen_album_url, gen_song_url, genius_album,
## genius_lyrics, genius_tracklist, genius_url, possible_album,
## possible_lyrics, prep_info
library(dplyr) #data manipulation
library(ggplot2) #visualizations
library(gridExtra) #viewing multiple plots together
##
## Attaching package: 'gridExtra'
## The following object is masked from 'package:dplyr':
##
## combine
library(tidytext) #text mining
library(wordcloud2) #creative visualizations
Sys.setenv(SPOTIFY_CLIENT_ID = '58f53eb1c69d4bd3a4aa2abd2ea265a1')
Sys.setenv(SPOTIFY_CLIENT_SECRET = '10b381cbeab548eb80d6f837be01e41e')
Sys.setenv(GENIUS_API_TOKEN = 'BOqFxFmHKNRwYEJuf0X5etaeaUhaZxNWiNGhYOqkvrLnh9STzd3hPgiYfvIv7w90')
accesstoken <- get_spotify_access_token()
# Spotify Artist : Foo Fighters
# spotify:artist:7jy3rLJdDQY21OgRLCZ9sD
#Extract data from spotify
ff_albums <- get_artist_albums(artist = 'Foo Fighters')
ff_meta <- get_albums(albums = ff_albums$album_uri)
ff_tracks <- get_album_tracks(ff_albums)
ff_audio_analysis <- data.frame(get_track_audio_analysis(ff_tracks$track_uri[1]))
ff_audio_features <- data.frame(get_track_audio_features(ff_tracks))
# using melt to reshape the columns in top_tracks into rows
# categories - danceability, energy, loudness, speechiness, acousticness, instrumentalness,
# liveness, valence, tempo
ff_audio_melt <- melt(ff_audio_features, id=(c("track_uri","duration_ms","time_signature","key_mode")))
# remove rows where variable = key or mode, not interested in those
ff_audio_melt <- ff_audio_melt %>%
filter(!(variable %in% c("key","mode")))
# set the value column to numeric
ff_audio_melt$value<-as.numeric(ff_audio_melt$value)
min_tempo <<- min(ff_audio_features$tempo)
min_dance <<- min(ff_audio_features$danceability)
min_energy <<- min(ff_audio_features$energy)
min_loudness <<- min(ff_audio_features$loudness)
min_speechiness <<- min(ff_audio_features$speechiness)
min_acoustic <<- min(ff_audio_features$acousticness)
min_instru <<- min(ff_audio_features$instrumentalness)
min_valence <<- min(ff_audio_features$valence)
min_live <<- min(ff_audio_features$liveness)
# max's
max_tempo <<- max(ff_audio_features$tempo)
max_dance <<- max(ff_audio_features$danceability)
max_energy <<- max(ff_audio_features$energy)
max_loudness <<- max(ff_audio_features$loudness)
max_speechiness <<- max(ff_audio_features$speechiness)
max_acoustic <<- max(ff_audio_features$acousticness)
max_instru <<- max(ff_audio_features$instrumentalness)
max_valence <<- max(ff_audio_features$valence)
max_live <<- max(ff_audio_features$liveness)
# function for normalising the data, formula => (value - min) / (max - min)
norm_data <- function(var1,x_val) {
if (var1=="tempo") {
norm_val <- (x_val-min_tempo)/(max_tempo-min_tempo)
}
if (var1=="liveness") {
norm_val <- (x_val-min_live)/(max_live-min_live)
}
if (var1=="danceability") {
norm_val <- (x_val-min_dance)/(max_dance-min_dance)
}
if (var1=="energy") {
norm_val <- (x_val-min_energy)/(max_energy-min_energy)
}
if (var1=="loudness") {
norm_val <- (x_val-min_loudness)/(max_loudness-min_loudness)
}
if (var1=="speechiness") {
norm_val <- (x_val-min_speechiness)/(max_speechiness-min_speechiness)
}
if (var1=="acousticness") {
norm_val <- (x_val-min_acoustic)/(max_acoustic-min_acoustic)
}
if (var1=="instrumentalness") {
norm_val <- (x_val-min_instru)/(max_instru-min_instru)
}
if (var1=="valence") {
norm_val <- (x_val-min_valence)/(max_valence-min_valence)
}
return(norm_val)
}
# normalise the audio values for the tt_audio_melt dataframe
for (row in 1:nrow(ff_audio_melt)) {
# print(tt_audio_melt$value)
ff_audio_melt$norm_value[row] <- norm_data(ff_audio_melt$variable[row],ff_audio_melt$value[row])
}
ff_sorted <- ff_audio_melt %>%
group_by(variable) %>%
summarise(m = median(norm_value)) %>%
arrange(m) %>%
.[["variable"]]
# Create Boxplot for audio features on top tracks, data is normalised
p <- plot_ly(data = ff_audio_melt, x = ~factor(variable,ff_sorted), y = ~norm_value, type = "box",
color = ~variable, boxpoints = "all", jitter = 0.3, pointpos = -1.8)
p
## Warning in RColorBrewer::brewer.pal(N, "Set2"): n too large, allowed maximum for palette Set2 is 8
## Returning the palette you asked for with that many colors
## Warning in RColorBrewer::brewer.pal(N, "Set2"): n too large, allowed maximum for palette Set2 is 8
## Returning the palette you asked for with that many colors
# Lyrics Analysis - utilising GeniusR library
song_ids <- c()
no_data<-"N"
for (row in 1:nrow(ff_tracks)) {
no_data<-"N"
tryCatch({
lyrics <- genius_lyrics(artist = "Foo Fighters", song = ff_tracks$track_name[row])
},error=function(e) {
no_data<-"Y"},
finally = {
if (no_data=="N") {
if (nrow(lyrics)>0) {
if (row==1) {
lyrics$song_id <- row
lyrics_ds <- lyrics
}
if (row>1) {
lyrics$song_id <- row
lyrics_ds <- rbind(lyrics_ds,lyrics)
}
lyrics <- lyrics[0,]
song_ids <- c(song_ids, row)
}
}
no_data<-"N"})
}
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
## Warning in request_GET(session, url): Not Found (HTTP 404).
# Sentiment Analysis
l_analyse <- lyrics_ds %>%
select(lyric,song_id) %>%
group_by(song_id) %>%
unnest_tokens(word,lyric)
nrc_joy <- get_sentiments("nrc") %>%
filter(sentiment == "joy")
nrc_anger <- get_sentiments("nrc") %>%
filter(sentiment == "anger")
nrc_sent <- get_sentiments("nrc")
afinn_sent <- get_sentiments("afinn")
bing_sent <- get_sentiments("bing")
# look for joy words
words_joy <- l_analyse %>%
inner_join(nrc_joy) %>%
count(word, sort = TRUE)
## Joining, by = "word"
# look for anger words
words_anger <- l_analyse %>%
inner_join(nrc_anger) %>%
count(word, sort = TRUE)
## Joining, by = "word"
total_words <- rbind(words_joy,words_anger)
# analysing the JOY words
s_analysis <- total_words %>%
inner_join(nrc_sent)
## Joining, by = "word"
sentiments_full <- s_analysis %>%
count(sentiment, sort = TRUE) %>%
arrange(song_id,sentiment)
s_analysis_bing <- total_words %>%
inner_join(bing_sent)
## Joining, by = "word"
sentiments_bing <- s_analysis_bing %>%
count(sentiment, sort = TRUE) %>%
arrange(song_id,sentiment)
s_analysis_afinn <- total_words %>%
inner_join(afinn_sent)
## Joining, by = "word"
sentiments_afinn <- s_analysis_afinn %>%
count(score, sort = TRUE) %>%
arrange(song_id,score)
# Plots of Sentiment Analysis
cbbPalette <- c("#000000", "#E69F00", "#56B4E9", "#009E73", "#F0E442", "#0072B2", "#D55E00", "#CC79A7")
# To use for fills, add
scale_fill_manual(values=cbPalette)
## <ggproto object: Class ScaleDiscrete, Scale, gg>
## aesthetics: fill
## axis_order: function
## break_info: function
## break_positions: function
## breaks: waiver
## call: call
## clone: function
## dimension: function
## drop: TRUE
## expand: waiver
## get_breaks: function
## get_breaks_minor: function
## get_labels: function
## get_limits: function
## guide: legend
## is_discrete: function
## is_empty: function
## labels: waiver
## limits: NULL
## make_sec_title: function
## make_title: function
## map: function
## map_df: function
## n.breaks.cache: NULL
## na.translate: TRUE
## na.value: NA
## name: waiver
## palette: function
## palette.cache: NULL
## position: left
## range: <ggproto object: Class RangeDiscrete, Range, gg>
## range: NULL
## reset: function
## train: function
## super: <ggproto object: Class RangeDiscrete, Range, gg>
## reset: function
## scale_name: manual
## train: function
## train_df: function
## transform: function
## transform_df: function
## super: <ggproto object: Class ScaleDiscrete, Scale, gg>
ggplot(sentiments_full, aes(sentiment, nn, fill = sentiment)) +
geom_col() +
theme(axis.text.x = element_text(angle = 45, hjust = 1, face="bold")) +
theme(
plot.title = element_text(color="red", size=9, face="bold.italic"),
axis.title.x = element_text(color="black", size=9, face="bold"),
axis.title.y = element_text(color="black", size=9, face="bold")
)
ggplot(sentiments_afinn, aes(score, nn, fill = score)) +
geom_col() +
theme(axis.text.x = element_text(angle = 45, hjust = 1, face="bold")) +
theme(
plot.title = element_text(color="red", size=9, face="bold.italic"),
axis.title.x = element_text(color="black", size=9, face="bold"),
axis.title.y = element_text(color="black", size=9, face="bold")
)
ggplot(sentiments_bing, aes(sentiment, nn, fill = sentiment)) +
geom_col() +
theme(axis.text.x = element_text(angle = 45, hjust = 1, face="bold")) +
theme(
plot.title = element_text(color="red", size=9, face="bold.italic"),
axis.title.x = element_text(color="black", size=9, face="bold"),
axis.title.y = element_text(color="black", size=9, face="bold")
)
# Most common positive and negative words
bing_word_counts <- l_analyse %>%
inner_join(get_sentiments("bing")) %>%
count(word, sentiment, sort = TRUE) %>%
ungroup()
## Joining, by = "word"
bing_word_counts %>%
group_by(sentiment) %>%
top_n(5) %>%
ungroup() %>%
mutate(word = reorder(word, n)) %>%
ggplot(aes(word, n, fill = sentiment)) +
geom_col(show.legend = FALSE) +
facet_wrap(~sentiment, scales = "free_y") +
labs(y = "Contribution to sentiment",
x = NULL) +
coord_flip()
## Selecting by n
twords <- total_words[c('word','n')]
twords %>%
anti_join(stop_words) %>%
count(word) %>%
wordcloud2(word, size=1.6, color='random-light', backgroundColor="black")
## Joining, by = "word"